# Towards Exploiting Geometry and Time for Fast Off-Distribution Adaptation in Multi-Task Robot Learning

arXiv: https://arxiv.org/abs/2106.13237 · HTML: https://mlsanity.com/paper/2106.13237/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 17 references that are in the map.
- First posted on arXiv 24 Jun 2021.

## Builds on (most cited first, 11 of 11)

- [OpenAI Gym](https://mlsanity.com/paper/1606.01540/) (2016)
- [QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation](https://mlsanity.com/paper/1806.10293/) (2018)
- [Data-Efficient Hierarchical Reinforcement Learning](https://mlsanity.com/paper/1805.08296/) (2018)
- [Relay Policy Learning: Solving Long-Horizon Tasks via Imitation and Reinforcement Learning](https://mlsanity.com/paper/1910.11956/) (2019)
- [Residual Policy Learning](https://mlsanity.com/paper/1812.06298/) (2018)
- [MDP Homomorphic Networks: Group Symmetries in Reinforcement Learning](https://mlsanity.com/paper/2006.16908/) (2020)
- [Language as an Abstraction for Hierarchical Deep Reinforcement Learning](https://mlsanity.com/paper/1906.07343/) (2019)
- [Plannable Approximations to MDP Homomorphisms: Equivariance under Actions](https://mlsanity.com/paper/2002.11963/) (2020)
- [SE3-Nets: Learning Rigid Body Motion using Deep Neural Networks](https://mlsanity.com/paper/1606.02378/) (2016)
- [Composable Planning with Attributes](https://mlsanity.com/paper/1803.00512/) (2018)
- [GRAC: Self-Guided and Self-Regularized Actor-Critic](https://mlsanity.com/paper/2009.08973/) (2020)

## Abstract (arXiv)

We explore possible methods for multi-task transfer learning which seek to exploit the shared physical structure of robotics tasks. Specifically, we train policies for a base set of pre-training tasks, then experiment with adapting to new off-distribution tasks, using simple architectural approaches for re-using these policies as black-box priors. These approaches include learning an alignment of either the observation space or action space from a base to a target task to exploit rigid body structure, and methods for learning a time-domain switching policy across base tasks which solves the target task, to exploit temporal coherence. We find that combining low-complexity target policy classes, base policies as black-box priors, and simple optimization algorithms allows us to acquire new tasks outside the base task distribution, using small amounts of offline training data.